An automatic evaluation method and system for the curative effect of nasopharyngeal carcinoma

Abstract

The invention discloses an automatic evaluation method and system for the curative effect of nasopharyngeal carcinoma. The method includes: using a deep learning algorithm to segment nasopharyngeal carcinoma images to obtain nasopharyngeal carcinoma lesions before treatment and nasopharyngeal carcinoma lesions after treatment; The post-treatment nasopharyngeal carcinoma lesions are registered to the pre-treatment nasopharyngeal carcinoma lesions, and the registered nasopharyngeal carcinoma lesions are obtained; according to the pre-treatment nasopharyngeal carcinoma lesions and the registered post-treatment nasopharyngeal carcinoma lesions, an automatic efficacy evaluation algorithm is used Carry out automatic curative effect evaluation and output curative effect evaluation results. The present invention makes the image of nasopharyngeal carcinoma after treatment and the image of nasopharyngeal carcinoma before treatment in the same image space through registration, which is more conducive to accurately calculating the difference of lesions before and after treatment. Segmentation can automatically segment lesions from the original nasopharyngeal carcinoma image, which is more accurate; the automatic curative effect evaluation algorithm is used to automatically evaluate the curative effect, and the curative effect evaluation result can be automatically obtained, which is faster and more accurate, and can be widely used in the field of medical computer applications.

Description

technical field [0001] The invention relates to the application field of medical computers, in particular to an automatic evaluation method and system for the curative effect of nasopharyngeal carcinoma. Background technique [0002] Nasopharyngeal carcinoma is a malignant tumor. The incidence of nasopharyngeal cancer has significant geographical differences. China is in the high incidence area of ​​nasopharyngeal cancer, of which Guangdong is the highest incidence area, so nasopharyngeal cancer is also known as "Cantonese cancer". Currently, radiation therapy is the most effective treatment for nasopharyngeal carcinoma. During the treatment of nasopharyngeal cancer, the shape and location of the tumor will change, large or small, as the course of radiation therapy progresses. Doctors need to evaluate the treatment effect according to the changes before and after the tumor treatment, so as to make the next radiotherapy plan more accurately. [0003] At present, doctors wi...

AI Technical Summary

Problems solved by technology

[0003] At present, doctors will analyze the clinical images of patients before and after radiotherapy, and evaluate the curative effect of this treatment according to the recist 1.1 standard. Efficiency: 1) The pre-treatment image and the post-treatment image are not imaged at the same time, and the operating technician may not be the same person. Even in the same hospital, there will be multiple imaging machines, which cannot guarantee the consistency of the pre-treatment image and the post-treatment image

2) Doctors with different experience will have differences in locating nasopharyngeal carcinoma lesions, and doctors with less experience may not be accurate enough in locating the lesion area

3) When evaluating the curative effect, the doctor needs to compare the pre-treatment and post-treatment images layer by layer with the naked eye, which takes a lot of time and is not accurate enough

Images